Only if the GSH is dirty

Thanks for Leopold Flohe's answer. I don't thinks this problem is cased by the impurities of GSH, the GSH was bought from sigma recently and the purity over 98%, besides, this flocculent flotage was never appear when I dialysis in PBS buffer.

Please see the paper entitled 

Studies on Glutathione S-Alkyltransferase of the Rat

By M. K. JOHNSON

Toxicology Research Unit, Medical Research Council Laboratories,

Woodmansterne Road, Carshalton, Surrey

published in Biochem. J. (1966) 98, 44. Please check the activity of the enzyme purified by you through Sepharose 4B-GST tag

See page 47 where reaction of the purified enzyme with NaCl is described . Briefly I have quoted here :  

Dialysis of liver extracts against deionized water

for 5hr. at 2° caused a heavy flocculent precipitate

to form, and only 28% of the activity was recovered

from the soluble protein and almost none from the

precipitate, which was mostly insoluble in 2%

sodium chloride. A similar effect was observed with

kidney extract. This loss of activity was due to

denaturation of the protein in the solution of low

ionic strength rather than to loss of cofactor, since

dialysis for the same time against 2% sodium

chloride solution in deionized water led to only a

slight precipitation of protein and retention of 72-

95% of the activity in solution in several experiments.

I do hope this helps to understand the flocculation.

Thanks for Shelley Bhattacharya's help,  I will change the concentration of sodium chloride to check the effect of ion concentration.